PANIPREP's toolbox

Expanding and exploiting our knowledge of the ‘antivirome’

PANVIPREP will combine conventional methods for drug discovery and development with a number of innovative approaches and improved methodologies, in particular: 

A three-track approach to identify and improve antiviral hits

To identify and improve potent drug-like compounds, we will employ three parallel approaches in work packages 2, 3 and 4:

• WP1 - From hit to early lead: development of selected hit series
• WP2 - From hit to new target: target identification and evaluation by exploratory chemistry & innovative drug design approaches
• WP3 - From target to new hits: exploring established viral drug targets to discover innovative drug approaches

This integrated approach closes the drug discovery loop and will deliver better starting points for future drug development. Simultaneously, novel technological approaches will be incorporated in a robust pipeline for broad-spectrum antiviral drug development. Data files will be formatted to enable AI-based approaches involving PANVIPREP partners from multiple scientific disciplines. 

Novel viral drug targets and drug candidates

Compound series will be considered novel potential drug candidates if they are structurally distinct from existing compounds and active against a relevant target that is not covered by existing patents. We will use compound series for further development if a novel binding site or mode-of-action can be revealed. PANVIPREP may thus deliver broad-spectrum hits with significant novelty.

Artificial Intelligence and Machine Learning to perform hit enrichment

This approach aims to radically reduce the time and cost of new hit discovery. Building a model trained to perform a supervised analysis on pre-existing data sets can identify novel chemistry in virtual libraries, which are more likely to be active antiviral hits. Connections to AI-based structural models (e.g., AlphaFold2) will be established.

Using PROTACs to degrade viral target proteins specifically

PROTACS trigger viral target degradation through the cellular proteasome and do not rely on stochiometric binding. This approach has been used in cancer research, but has remained largely unexplored in the antiviral drug field. We will use innovative designs and methods to demonstrate target engagement, and improve our methodology for PROTAC design and characterization. 

In silico PK analysis to identify compounds with favorable in vivo characteristics

Compound classes will be screened in silico to identify the best candidates early in hit-to-lead development and preview their suitability for testing in animal models. 

Developing biosafe models to enhance screening throughput

Infection models relying on the use of engineered biosafe models will be used to speed up the screening process. Selected viruses will be attenuated by using large-scale genome recoding, and we will employ non-infectious replicon systems and single-round infectious viruses to facilitate mechanism-of-action studies under BSL2 conditions.